Duplicate systems in HP Insight Manager 6.3

[lambdaxy]

Hello,

For some time now I've had duplicate systems registered on the CMS. Here is an example:

Instance 1 (device is not associated to an enclosure and has no ip address.)

Preferred System Name ************

Network Name ***********************************

UUID 35333434-3732-4247-3837-353035485636

Product Description

System Type Server

System Subtype HP ProLiant, Blade, HP BladeSystem c-Class, x86 System, Virtual Machine Host, VMware ESX Host Product Model ProLiant BL680c G5

Instance 2 (Device the same with same preferred name and network name is completely knwon)

Identification

Address **********

Preferred System Name **********

Network Name ******************************

Serial Number GB87505HV6

Product Number 443527-B21 UUID 35333434-3732-4247-3837-353035485636

Product Description

System Type Server

System Subtype HP ProLiant, Blade, HP BladeSystem c-Class, x86 System, Virtual Machine Host, VMware ESX Host Product Model ProLiant BL680c G5

Hardware Description Linux SCTV0049 2.6.18-274.ESX #1 Tue Feb 14 18:09:36 PST 2012 x86_64 OS

Name VMware ESX

OS Type Hypervisor OS Version 4.1.0 OS

Description VMware ESX 4.1.0 build-702113

Management Protocols HTTP:, SNMP:1.0, WBEM:1.0, SSH:SSH-2.0-OpenSSH_4.3, SMH:2.0

Even if I keep deleteing the instance of the device that is not associated to the enclosure it keeps coming back.

Any idea would be much appreciated since I am a bit puzzled.

[Tushar Bajpai]

Download and install the latest hotfix for HPSIM 6.3. The issue is address in that.

ftp://ftp.hp.com/pub/softlib2/software1/pubsw-windows/p67295809/v79998

Thanks

[lambdaxy]

Thanks for the reply.

Unfortunately all updates up to "HOTFIX63_015.jar wo, 14/03/2012, 16:01 CET" are already installed on the system.

[Hafedh_K]

Hello,

I'm facing the same issue, and all patches were already installed , but this doesn't resolve the issue....

Do you have another action plan plz ?

[lambdaxy]

Hello,

Nobody has got any more information on this particular problem?

A ticket support has been open for several weeks already but no solution yet. So far they only proposed:

• to delete the duplicates and rediscover them
• to uncheck the option to discover automatically new servers when iLO is discovered.

Both options were unsuccessfull.

[shocko]

COuple of things to check:

1. Check that all systems involved have a unique Serial number in their BIOS and also that they have a product ID set in the BIOS
2. Check that the iLO is reading these values
3. Check in the system properties in SIM that no manual entries have been configured for system name, IP address, serial number/product ID and that the check box to prevent discovery overwriting these values is not checked.

[lambdaxy]

Hello,

Thank you for the feedback.

Almost all Blade servers are concerned. While I did not check them all I am confident they all have a unique serial number and product ID and am confident as well that SIM knows about it since we have automated processes that uses them.

I check if some system properties were locked on a problematic blade server. Here is a partial result of the mxnode -ld command:

System name locked: No
System attributes locked: No
System Primary IP locked: No

For some reason the duplicates were created at some point. However it seems that the software cannot recover from the error... because it uses the system name for many of its queries and that system name is duplicated.

For instance if I delete on of the instance of a duplicate server and try subsequently to identify the remaining one I recieve an error message indicating that the server has been deleted. Also the delete of both systems and rediscovery does not work. It is as if the delete is played in some sort of cache and not in the database itself. Upon a restart of the service the deleted object comes back.